Introduction to Computational Linguistics

Introduction to Computational Linguistics
Winter 2010/2011

Course description:

This course provides a non-technical introduction to the field of computational linguistics and its history. We will cover major application areas of computational linguistics including machine translation, information retrieval, information extraction, and computational lexicography. We will also discuss the tools and resources needed for natural language processing and generation.

Course information:

Meeting time: Wednesday 2:15-3:45pm, SfS Hörsaal 0.02
Instructor: Adriane Boyd (adriane AT sfs DOT uni-tuebingen DOT de)
Website: http://www.sfs.uni-tuebingen.de/~adriane/teaching/ws2010/icl/
Course syllabus

Announcements

12/15: take a look at the revised schedule below!
No office hours on 12/23.
No office hours on 11/18.

Links

Named entity recognition demo:
- Named entity recognition demo with colorized output from the University of Illinois
Online part-of-speech tagger demos:
- POS tagger demo with colorized output from the University of Illinois (Penn treebank tagset)
- CLAWS tagger demo from the British National Corpus
  Enter any English text you want! It's best if you only try out shorter texts (a couple sentences), so you don't overload their server.
- If you want experiment more or tag a large text, download a tagger. Some good freely available taggers are: Stanford POS Tagger (English, German, Arabic, Chinese), TreeTagger (many languages), OpenNLP tools (English, German, Spanish, Thai), LingPipe tools (English for different genres)
A few online corpus search engines with POS-based queries:
- CQP Demos: Dickens (Penn treebank tagset) and Bundestag corpora (German, STTS tagset)
- TIME Magazine Corpus (CLAWS tagset)
- British National Corpus (CLAWS tagset)
Finite-state automata libraries and toolkits:
- AT&T Finite-State Machine Library (commandline tools; this is the only one I've used, it's used a lot in speech recognition)
- Gerd van Noord's Finite State Automata Utilities (prolog)
- dk.brics.automaton (java)
- OpenFST (C, just for weighted finite-state transducers)
Python Natural Language Toolkit (NLTK)
Zwiebelfisch: Ziehen Sie die Brille aus! (see the Zwiebelfisch archive for more)
Geoffrey K. Pullum on Eskimo snow vocabulary (Language Log is a very good linguistics blog with many well-known and well-regarded contributors)
Google 5-Gram Database Search (only from within the university network!)

Schedule

Month	Date	Topic	Readings
Oct	20	Introduction
	27	Historical Overview, Machine Translation	Locke and Booth (1955), Jurafsky and Martin (2008)
Nov	3	Machine Translation	Hutchins and Somers (1992), Somers (2000)
	10	Machine Translation	Knight (1997)
	17	Tokenization and Sentence Segmentation	Palmer (2000)
	24	Regular Expressions and Finite State Automata	Karttunen (2003)
Dec	1	Morphological Analysis and Finite State Transducers	Wintner (2010), Jurafsky and Martin (2009)
	8	Midterm Exam
	15	Part-of-Speech Tagging	Leech (1997), Jurafsky and Martin (2009)
	22	Part-of-Speech Tagging	(no new reading)
Jan	12	Corpus Construction and Annotation	Palmer and Xue (2010)
	19	Computational Lexicography	Jurafsky and Martin (2009), Chapters 19-20.3
	26	Information Extraction, Review	Jurafsky and Martin (2009), Chapter 22-22.2
Feb	2	Final Exam